Online multimodal compression using pruning and knowledge distillation for iris recognition
Deep learning models have advanced to the forefront of image recognition tasks, resulting in high-performing but enormous neural networks with millions to billions of parameters. Yet, deploying these models in production systems imposes considerable memory limits. Hence, the research community is...
Saved in:
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English English |
Published: |
Semarak Ilmu Sdn Bhd
2024
|
Subjects: | |
Online Access: | http://irep.iium.edu.my/111162/3/111162_Online%20multimodal%20compression%20using%20pruning_SCOPUS.pdf http://irep.iium.edu.my/111162/4/111162_Online%20multimodal%20compression%20using%20pruning.pdf http://irep.iium.edu.my/111162/ https://semarakilmu.com.my/journals/index.php/applied_sciences_eng_tech/index https://doi.org/10.37934/araset.37.2.6881 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.iium.irep.111162 |
---|---|
record_format |
dspace |
spelling |
my.iium.irep.1111622024-03-16T00:43:41Z http://irep.iium.edu.my/111162/ Online multimodal compression using pruning and knowledge distillation for iris recognition Abdul Latif, Samihah Sidek, Khairul Azami Abu Bakar, Eddyzulham Hassan Abdalla Hashim, Aisha TK7885 Computer engineering Deep learning models have advanced to the forefront of image recognition tasks, resulting in high-performing but enormous neural networks with millions to billions of parameters. Yet, deploying these models in production systems imposes considerable memory limits. Hence, the research community is increasingly aware of the need for compression strategies that can reduce the number of model parameters and their resource requirement. Current compression techniques for deep learning models have limitations in efficiency and effectiveness, indicating that more research is required to develop more efficient and practical techniques capable of balancing the trade-offs between compression rate, computational cost, and accuracy. This study proposed a multimodal method by combining multimodal Pruning and Knowledge Distillation techniques for compressing the iris recognition model, which is the size constraint for many image recognition models. To maintain accuracy while shrinking the model’s size, the models are trained, compressed, and further retrained in the downstream job. The analysis includes both fully connected and convolutional layers. Experimentally, the findings show that the proposed technique can achieve 91% accuracy, the same as the existing or original model. Besides that, the model compression can reduce the size of the model almost six times, from 529MB to 90MB, which is a significantly reduced rate. The primary outcome of this study is developing a CNN lightweight model for iris recognition technology that can be used on mobile devices and is resource constrained. Semarak Ilmu Sdn Bhd 2024-01-16 Article PeerReviewed application/pdf en http://irep.iium.edu.my/111162/3/111162_Online%20multimodal%20compression%20using%20pruning_SCOPUS.pdf application/pdf en http://irep.iium.edu.my/111162/4/111162_Online%20multimodal%20compression%20using%20pruning.pdf Abdul Latif, Samihah and Sidek, Khairul Azami and Abu Bakar, Eddyzulham and Hassan Abdalla Hashim, Aisha (2024) Online multimodal compression using pruning and knowledge distillation for iris recognition. Journal of Advanced Research in Applied Sciences and Engineering Technology, 37 (2). pp. 68-81. ISSN 2462-1943 https://semarakilmu.com.my/journals/index.php/applied_sciences_eng_tech/index https://doi.org/10.37934/araset.37.2.6881 |
institution |
Universiti Islam Antarabangsa Malaysia |
building |
IIUM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
International Islamic University Malaysia |
content_source |
IIUM Repository (IREP) |
url_provider |
http://irep.iium.edu.my/ |
language |
English English |
topic |
TK7885 Computer engineering |
spellingShingle |
TK7885 Computer engineering Abdul Latif, Samihah Sidek, Khairul Azami Abu Bakar, Eddyzulham Hassan Abdalla Hashim, Aisha Online multimodal compression using pruning and knowledge distillation for iris recognition |
description |
Deep learning models have advanced to the forefront of image recognition tasks,
resulting in high-performing but enormous neural networks with millions to billions of
parameters. Yet, deploying these models in production systems imposes considerable
memory limits. Hence, the research community is increasingly aware of the need for
compression strategies that can reduce the number of model parameters and their
resource requirement. Current compression techniques for deep learning models have
limitations in efficiency and effectiveness, indicating that more research is required to
develop more efficient and practical techniques capable of balancing the trade-offs
between compression rate, computational cost, and accuracy. This study proposed a
multimodal method by combining multimodal Pruning and Knowledge Distillation
techniques for compressing the iris recognition model, which is the size constraint for
many image recognition models. To maintain accuracy while shrinking the model’s size,
the models are trained, compressed, and further retrained in the downstream job. The
analysis includes both fully connected and convolutional layers. Experimentally, the
findings show that the proposed technique can achieve 91% accuracy, the same as the
existing or original model. Besides that, the model compression can reduce the size of
the model almost six times, from 529MB to 90MB, which is a significantly reduced rate.
The primary outcome of this study is developing a CNN lightweight model for iris
recognition technology that can be used on mobile devices and is resource constrained. |
format |
Article |
author |
Abdul Latif, Samihah Sidek, Khairul Azami Abu Bakar, Eddyzulham Hassan Abdalla Hashim, Aisha |
author_facet |
Abdul Latif, Samihah Sidek, Khairul Azami Abu Bakar, Eddyzulham Hassan Abdalla Hashim, Aisha |
author_sort |
Abdul Latif, Samihah |
title |
Online multimodal compression using pruning and knowledge distillation for iris recognition |
title_short |
Online multimodal compression using pruning and knowledge distillation for iris recognition |
title_full |
Online multimodal compression using pruning and knowledge distillation for iris recognition |
title_fullStr |
Online multimodal compression using pruning and knowledge distillation for iris recognition |
title_full_unstemmed |
Online multimodal compression using pruning and knowledge distillation for iris recognition |
title_sort |
online multimodal compression using pruning and knowledge distillation for iris recognition |
publisher |
Semarak Ilmu Sdn Bhd |
publishDate |
2024 |
url |
http://irep.iium.edu.my/111162/3/111162_Online%20multimodal%20compression%20using%20pruning_SCOPUS.pdf http://irep.iium.edu.my/111162/4/111162_Online%20multimodal%20compression%20using%20pruning.pdf http://irep.iium.edu.my/111162/ https://semarakilmu.com.my/journals/index.php/applied_sciences_eng_tech/index https://doi.org/10.37934/araset.37.2.6881 |
_version_ |
1794544806803800064 |
score |
13.18916 |